Kafka的高级消费者与低级消费者

          在Kafka实战章节，我们写的例子都是Kafka的高级消费实例，可以看到在消息消费者的程序中，我们只需要指定zookeeper、及消费群组的groupId即可实现从消息队列中消费消息，屏蔽了大量的底层细节：如消息的偏移量等信息都不在程序中维护。Kafka的高级消费实例，满足以下几点规则：

（1）同一个消费群组中，如果线程数大于Topic分区数，那么一些线程永远接收不到消息；

（2）同一个消费群组中，如果线程数小于Topic分区数，部分线程将从多个分区接收消息；

（3）对于从多个分区接收消息的线程，消费每个分区内的消息是有序的，但消费多个分区之间的消息是无序的；

        明白了Kafka的高级消费实例的过程之后，如果我们想进一步控制一个消费者消费哪个分区怎么办呢？比如多次读取同一个消息。答案是使用低级消费者实例，即在程序中指定Topic的Partition的Leader broker，并在程序中跟踪消息的偏移量offset值。其步骤大致如下：

（1）指定消费Topic Partition的Leader broker及备份broker；

（2）构造并发送请求数据；

（3）处理leader broker的变更；

        实例如下：

import kafka.api.FetchRequest;
import kafka.api.FetchRequestBuilder;
import kafka.api.PartitionOffsetRequestInfo;
import kafka.common.ErrorMapping;
import kafka.common.TopicAndPartition;
import kafka.javaapi.*;
import kafka.javaapi.consumer.SimpleConsumer;
import kafka.message.MessageAndOffset;

import java.nio.ByteBuffer;
import java.util.ArrayList;
import java.util.Collections;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

public class SimpleConsumerDemo {
    private List m_replicaBrokers = new ArrayList<>();

    public SimpleConsumerDemo(){
        m_replicaBrokers = new ArrayList<>();
    }

    public void run(long a_maxReads, String a_topic, int a_partition, List a_seedBrokers, int a_port) throws Exception {
        // find the meta data about the topic and partition we are interested in
        //
        PartitionMetadata metadata = findLeader(a_seedBrokers, a_port, a_topic, a_partition);
        if (metadata == null) {
            System.out.println("Can't find metadata for Topic and Partition. Exiting");
            return;
        }
        if (metadata.leader() == null) {
            System.out.println("Can't find Leader for Topic and Partition. Exiting");
            return;
        }
        String leadBroker = metadata.leader().host();
        String clientName = "Client_" + a_topic + "_" + a_partition;

        SimpleConsumer consumer = new SimpleConsumer(leadBroker, a_port, 100000, 64 * 1024, clientName);
        long readOffset = getLastOffset(consumer,a_topic, a_partition, kafka.api.OffsetRequest.EarliestTime(), clientName);

        int numErrors = 0;
        while (a_maxReads > 0) {
            if (consumer == null) {
                consumer = new SimpleConsumer(leadBroker, a_port, 100000, 64 * 1024, clientName);
            }
            FetchRequest req = new FetchRequestBuilder()
                    .clientId(clientName)
                    .addFetch(a_topic, a_partition, readOffset, 100000) // Note: this fetchSize of 100000 might need to be increased if large batches are written to Kafka
                    .build();
            FetchResponse fetchResponse = consumer.fetch(req);

            if (fetchResponse.hasError()) {
                numErrors++;
                // Something went wrong!
                short code = fetchResponse.errorCode(a_topic, a_partition);
                System.out.println("Error fetching data from the Broker:" + leadBroker + " Reason: " + code);
                if (numErrors > 5) break;
                if (code == ErrorMapping.OffsetOutOfRangeCode())  {
                    // We asked for an invalid offset. For simple case ask for the last element to reset
                    readOffset = getLastOffset(consumer,a_topic, a_partition, kafka.api.OffsetRequest.LatestTime(), clientName);
                    continue;
                }
                consumer.close();
                consumer = null;
                leadBroker = findNewLeader(leadBroker, a_topic, a_partition, a_port);
                continue;
            }
            numErrors = 0;

            long numRead = 0;
            for (MessageAndOffset messageAndOffset : fetchResponse.messageSet(a_topic, a_partition)) {
                long currentOffset = messageAndOffset.offset();
                if (currentOffset < readOffset) {
                    System.out.println("Found an old offset: " + currentOffset + " Expecting: " + readOffset);
                    continue;
                }
                readOffset = messageAndOffset.nextOffset();
                ByteBuffer payload = messageAndOffset.message().payload();

                byte[] bytes = new byte[payload.limit()];
                payload.get(bytes);
                System.out.println(String.valueOf(messageAndOffset.offset()) + ": " + new String(bytes, "UTF-8"));
                numRead++;
                a_maxReads--;
            }

            if (numRead == 0) {
                try {
                    Thread.sleep(1000);
                } catch (InterruptedException ie) {
                }
            }
        }
        if (consumer != null) consumer.close();
    }

    public static long getLastOffset(SimpleConsumer consumer, String topic, int partition,
                                     long whichTime, String clientName) {
        TopicAndPartition topicAndPartition = new TopicAndPartition(topic, partition);
        Map requestInfo = new HashMap();
        requestInfo.put(topicAndPartition, new PartitionOffsetRequestInfo(whichTime, 1));
        kafka.javaapi.OffsetRequest request = new kafka.javaapi.OffsetRequest(
                requestInfo, kafka.api.OffsetRequest.CurrentVersion(), clientName);
        OffsetResponse response = consumer.getOffsetsBefore(request);

        if (response.hasError()) {
            System.out.println("Error fetching data Offset Data the Broker. Reason: " + response.errorCode(topic, partition) );
            return 0;
        }
        long[] offsets = response.offsets(topic, partition);
        return offsets[0];
    }

    private String findNewLeader(String a_oldLeader, String a_topic, int a_partition, int a_port) throws Exception {
        for (int i = 0; i < 3; i++) {
            boolean goToSleep = false;
            PartitionMetadata metadata = findLeader(m_replicaBrokers, a_port, a_topic, a_partition);
            if (metadata == null) {
                goToSleep = true;
            } else if (metadata.leader() == null) {
                goToSleep = true;
            } else if (a_oldLeader.equalsIgnoreCase(metadata.leader().host()) && i == 0) {
                // first time through if the leader hasn't changed give ZooKeeper a second to recover
                // second time, assume the broker did recover before failover, or it was a non-Broker issue
                //
                goToSleep = true;
            } else {
                return metadata.leader().host();
            }
            if (goToSleep) {
                try {
                    Thread.sleep(1000);
                } catch (InterruptedException ie) {
                }
            }
        }
        System.out.println("Unable to find new leader after Broker failure. Exiting");
        throw new Exception("Unable to find new leader after Broker failure. Exiting");
    }

    private PartitionMetadata findLeader(List a_seedBrokers, int a_port, String a_topic, int a_partition) {
        PartitionMetadata returnMetaData = null;
        loop:
        for (String seed : a_seedBrokers) {
            SimpleConsumer consumer = null;
            try {
                consumer = new SimpleConsumer(seed, a_port, 100000, 64 * 1024, "leaderLookup");
                List topics = Collections.singletonList(a_topic);
                TopicMetadataRequest req = new TopicMetadataRequest(topics);
                kafka.javaapi.TopicMetadataResponse resp = consumer.send(req);

                List metaData = resp.topicsMetadata();
                for (TopicMetadata item : metaData) {
                    for (PartitionMetadata part : item.partitionsMetadata()) {
                        if (part.partitionId() == a_partition) {
                            returnMetaData = part;
                            break loop;
                        }
                    }
                }
            } catch (Exception e) {
                System.out.println("Error communicating with Broker [" + seed + "] to find Leader for [" + a_topic
                        + ", " + a_partition + "] Reason: " + e);
            } finally {
                if (consumer != null) consumer.close();
            }
        }
        if (returnMetaData != null) {
            m_replicaBrokers.clear();
            for (kafka.cluster.Broker replica : returnMetaData.replicas()) {
                m_replicaBrokers.add(replica.host());
            }
        }
        return returnMetaData;
    }

    public static void main(String args[]) {
        SimpleConsumerDemo example = new SimpleConsumerDemo();
        long maxReads = Long.parseLong(args[0]);
        String topic = args[1];
        int partition = Integer.parseInt(args[2]);
        List seeds = new ArrayList<>();
        seeds.add(args[3]);
        int port = Integer.parseInt(args[4]);
        try {
            example.run(maxReads, topic, partition, seeds, port);
        } catch (Exception e) {
            System.out.println("Oops:" + e);
            e.printStackTrace();
        }
    }
}

参考资料：

1、https://cwiki.apache.org/confluence/display/KAFKA/Index

2、http://www.nohup.cc/article/195/

3、http://blog.csdn.net/honglei915/article/details/37563647

4、http://orchome.com/11